About
About this project
This application is analyzing twitter data retrieved on December 14, 2022. It contains tweets from the 12/08/20222 to the 12/14/2022, from all over the world in 35 different languages. The tweets analyzed contain the hashtag #tesla. This hashtag was chosen because of recent news about Elon Musk and his fortune, recent takeover of Twitter and news about the Tesla stocks, next to few other things.
This application allows visitors to explore the data interactively.
The dashboards in this application will give different insights about the moments in time the hashtag was especially popular. Assumptions about the reasons for this will be made. Moreover, this application will show related hashtags and topics on Twitter. Eventually, the application contains an interactive map, which shows the geolocations of the tweets containing the hashtag #tesla.
KEY FINNDINGS: In the given timespan, Tesla was in the press a couple of times. First reason for this is that the Tesla stock have been crashing, but also skyrocketing during that time. Furthermore, Tesla’s CEO Elon Musk lost his title as “richest man alive” more than once to the Frenchman Bernard Arnault. As we can see in the wordcloud and hashtag network, #tesla ist often tweeted in combination with words like “@elonmusk”, “#elonmusk”, “tesla”, “spacex” and “teslaradar”. The fact that Elon Musk has been mentioned often supports the assumption that the hashtag has been used a lot because of the news around his person. Him being the CEO of SpaceX also support it. Yet, why “teslaradar” is bein mentioned often remains unclear due to missing news articles. For the hashtag network, I identified networks around Elon Musk close to the #tesla hashtag, which connected to the stockmarket and to Silicon Valley companies. Other networks related consulting firms and data, renewable energy or the tesla models. As for the geomap, we can observe that most tweets come from the US and Europe. Few tweets were tweeted in Australia, India, Southeast Asia, Japan and Latinamerica and the Carrabean. No tweets came from Africa or the Middle East.
POSSIBLE LIMITATIONS OF ANALYSIS: The first limitation encountered were the tweets that weren’t related to any language, since these tweets only contain generic hashtags, Yet, they make up for 453 tweets in total, which could influence the graphs significantly. Second, the short timespan of the data collecting makes it hard to identify patterns in search trends, or prevents the graphs to strike out. Last but not least, the wordcloud contained many names of European countries. Yet, the analysis did not make appearant, why they were mentioned that often. Perhaps a topic model would help with that question, or collecting tweets over a longer amount of time.
The application is built with the Shiny framework for the R programming language. The application layout is produced with the flexdashboard package, and the charts and maps use Leaflet.js, Highcharts, and ggplot2, all accessed through their corresponding R packages.
To learn more about my work, you can always contact me on dfalk@umass.edu.
Tweets over time
Tweets from 12/08/2022 - 12/14/2022
This chart shows the frequency of the hashtag #tesla used in the timespan from 12/08/20222 to 12/14/2022.
As displayed, the frequency was highest on December 12, 2022.
Tweets by language
This interactive chart shows again on which days between the 12/08/20222 and the 12/14/2022 the frequency of the usage of #tesla was the highest. Yet, this chart allows the user to explore the frequencies by language in which the tweets were published.
Related topics
Row
Wordcloud
This wordcloud shows the 100 most used words withtin all the tweets collected. The bigger a word appears, the more often it was used in tweets.
It becomes appearant that the words “#tesla”, “@elonmusk”, “tesla” and “#elonmusk” are particularly popular. This could have something to do with the ups and downs in the Tesla stocks and news around Elon Musk.
Hashtag networks
A hashtag network shows the connections a specific hashtag has to others.
In this case, the hastag “#tesla” was used.
Those can differ in their related topics.
As the user can observe in this network,
100 different hashtags were picked to form a bigger hashtag network.
Topics to be identified in this network are
the stockmarket
Silicon Valley companies
consulting firms and data
renewable energy
and the tesla models.
Geomap
Map with tweets containing the hashtag #tesla
This interactive map shows the geolocations to the tweets collected. When clicking on the pins on the map, the user will be able to read the text to the corresponding tweet.